The present invention relates to compositions that are useful as catalyst activators for olefin polymerizations. More particularly the present invention relates to such compositions that are particularly adapted for use in the coordination polymerization of unsaturated compounds having improved activation efficiency and performance. Such compositions are particularly advantageous for use in a polymerization process wherein catalyst, catalyst activator, and at least one polymerizable monomer are combined under polymerization conditions to form a polymeric product.
It is previously known in the art to activate Ziegler-Natta polymerization catalysts, particularly such catalysts comprising Group 3-10 metal complexes containing delocalized .pi.-bonded ligand groups, by the use of an activator. Generally in the absence of such an activator compound, also referred to as a cocatalyst, little or no polymerization activity is observed.
A class of suitable activators are Lewis acids, especially alumoxanes, which are generally believed to be oligomeric or polymeric alkylaluminoxy compounds, including cyclic oligomers. Examples of alumoxanes (also known as aluminoxanes) include methylalumoxane (MAO) made by hydrolysis of trimethylaluminum as well as modified methylalumoxane (MMAO), wherein a portion of the trimethylaluminum is replaced by a higher alkyl aluminum compound such as triisobutylaluminum. MMAO advantageously is more soluble in aliphatic solvents than is MAO.
Generally alumoxanes contain, on average about 1.5 alkyl groups per aluminum atom, and are prepared by reaction of trialkylaluminum compounds or mixtures of compounds with water (Reddy et al, Prog. Poly. Sci., 1995, 20, 309-367). The resulting product is in fact a mixture of various substituted aluminum compounds including especially, trialklyaluminum compounds (resulting from incomplete reaction of the trialkylaluminum starting reagent or decomposition of the alumoxane). The amount of such free trialkylaluminum compound in the mixture generally varies from 1 to 50 percent by weight of the total product.
Although effective in forming an active olefin polymerization catalyst when combined with a variety of Group 3-10 metal complexes, especially Group 4 metal complexes, generally a large excess of alumoxane compared to metal complex, such as, molar ratios from 100:1 to 10,000:1, is required in order to produce adequate rates of polymerization. Unfortunately, the use of such large excesses of cocatalyst is expensive and also results in polymer having an elevated residual aluminum content. This latter factor may adversely affect polymer properties, especially clarity and dielectric constant.
A different type of activator compound is a Bronsted acid salt capable of transferring a proton to form a cationic derivative or other catalytically active derivative of such Group 3-10, 13 metal complex, cationic charge transferring compounds, or cationic oxidizing activators, referred to collectively hereinafter as cationic activators. Preferred cationic activators are ammonium, sulfonium, phosphonium, oxonium, ferrocenium, silver, lead, carbonium or silylium compounds containing a cation/anion pair that is capable of rendering the Group 3-10 metal complex catalytically active. Preferred anions associated with this cation comprise fluorinated arylborate anions, more preferably, the tetrakis(pentafluorophenyl)borate anion. Additional suitable anions include sterically shielded, bridged diboron anions. Examples of such cationic activators are disclosed in U.S. Pat. Nos. 5,198,401, 5,132,380, 5,470,927, 5,153,157, 5,350,723, 5,189,192, 5,626,087 and in U.S. Pat. No. 5,447,895, the teachings of which are herein incorporated by reference.
Further suitable activators for activating metal complexes for olefin polymerization include neutral Lewis acids such as tris(perfluorophenyl)borane and tris(perfluorobiphenyl)borane. The former composition has been previously disclosed for the above stated end use in U.S. Pat. No. 5,721,185, and elsewhere, whereas the latter composition is disclosed in Marks, et al, J. Am. Chem. Soc. 1996, 118, 12451-12452. Additional teachings of the foregoing activators may be found in Chen, et al, J. Am. Chem. Soc. 1997, 119, 2582-2583, Jia et al, Organometallics, 1997, 16, 842-857, and Coles et al, J. Am. Chem. Soc. 1997, 119, 8126-8126.
In U.S. Pat. No. 5,453,410, a strong Lewis acid activator, especially tris(pentafluorophenyl)borane, was disclosed for use in combination with constrained geometry metal complexes in combination with an alumoxane. This combination beneficially resulted in effective catalyst activation at molar ratios of alumoxane to catalyst that are much lower than required in the absence of the Lewis acid. Suitably, molar ratios from 1:1 to 50:1 could be employed. In U.S. Pat. Nos. 5,527,929, 5,616,664, 5,470,993, 5,556,928, 5,624,878, the combination of an alumoxane and a strong Lewis acid such as tris(pentafluorophenyl)boron was disclosed as a suitable activator for use with the metal complexes therein disclosed wherein the metal was in the +2 formal oxidation state. It is known that an exchange reaction between aluminum trialkyl compounds and tris(perfluorophenyl)borane occurs under certain conditions. This phenomenon has been previously described in U.S. Pat. No. 5,602,269.
Tris(perfluorophenyl)aluminum is a strong Lewis acid as well. However, it generally performs poorly by itself as an activator compared with tris-(perfluorophenyl)borane at a 1:1 molar ratio with a metal complex. Similarly, It has further been demonstrated that active catalysts resulting from the use of an aluminate anion based upon tris-(perfluorophenyl)aluminum for the activation of ansa-metallocenes and biscyclopentadienyl derivatives of zirconium(IV) are generally of lower activity than those formed by the corresponding borane (Ewen, Stud. in Surf. Sci. Catal. 1994, 89, 405-410). A possible explanation for the poor performance of tris-(perfluorophenyl)aluminum as an activator for metallocenes involving a back exchange reaction of a perfluorophenyl group has been proposed by Bochmann et al (ACS Dallas Meeting, March 1998, Abs. number INOR 264, subsequently published, Organometallics, 1998, 17, 5908-5912).
In light of these apparent deficiencies, it would be desirable to provide activator compositions based on Lewis acids for activation of metal complexes, especially complexes of metals of Group 4 of the Periodic Table of the elements having improved ease of use, cocatalyst properties and efficiency.